Discussion:
[Numpy-discussion] ANN: Numpy 1.10.0b1 release
Charles R Harris
2015-08-11 21:23:08 UTC
Permalink
Hi All,

give this release a whirl and report any problems either on the
numpy-discussion list or by opening an issue on github.
I'm pleased to announce the first beta release of Numpy 1.10.0. There is
over a year's worth of enhancements and bug fixes in the 1.10.0 release, so
please give this release a whirl and report any problems either on the
numpy-discussion list or by opening an issue on github. Tarballs,
installers, and release notes may be found in the usual place at Sourceforge
<https://sourceforge.net/projects/numpy/files/NumPy/1.10.0b1/>.

Chuck
Nathan Goldbaum
2015-08-11 21:44:45 UTC
Permalink
Maybe this is just me, I tried to build the tarball I got from sourceforge
in a fresh virtualenv on my mac and received the following error:

clang: numpy/core/src/multiarray/buffer.c
clang: src/multiarray/cblasfuncs.c
clang: error: no such file or directory: 'src/multiarray/cblasfuncs.c'
clang: error: no input files
clang: error: no such file or directory: 'src/multiarray/cblasfuncs.c'
clang: error: no input files

Indeed, that file is not present in the tarball. My download of the tarball
from sourceforge has sha1 hash 424bcee49507a655260068e767bd198094fc5604.

It looks like the tarball from github (
https://github.com/numpy/numpy/archive/v1.10.0b1.tar.gz) has the needed
file.
Post by Charles R Harris
Hi All,
give this release a whirl and report any problems either on the
numpy-discussion list or by opening an issue on github.
I'm pleased to announce the first beta release of Numpy 1.10.0. There is
over a year's worth of enhancements and bug fixes in the 1.10.0 release, so
please give this release a whirl and report any problems either on the
numpy-discussion list or by opening an issue on github. Tarballs,
installers, and release notes may be found in the usual place at
Sourceforge <https://sourceforge.net/projects/numpy/files/NumPy/1.10.0b1/>
.
Chuck
_______________________________________________
NumPy-Discussion mailing list
http://mail.scipy.org/mailman/listinfo/numpy-discussion
Charles R Harris
2015-08-11 22:13:23 UTC
Permalink
Post by Nathan Goldbaum
Maybe this is just me, I tried to build the tarball I got from sourceforge
clang: numpy/core/src/multiarray/buffer.c
clang: src/multiarray/cblasfuncs.c
clang: error: no such file or directory: 'src/multiarray/cblasfuncs.c'
clang: error: no input files
clang: error: no such file or directory: 'src/multiarray/cblasfuncs.c'
clang: error: no input files
Indeed, that file is not present in the tarball. My download of the
tarball from sourceforge has sha1
hash 424bcee49507a655260068e767bd198094fc5604.
It looks like the tarball from github (
https://github.com/numpy/numpy/archive/v1.10.0b1.tar.gz) has the needed
file.
Hmm, interesting. I'll take a look.

<snip>

Chuck
Charles R Harris
2015-08-11 22:22:13 UTC
Permalink
Post by Charles R Harris
Post by Nathan Goldbaum
Maybe this is just me, I tried to build the tarball I got from
sourceforge in a fresh virtualenv on my mac and received the following
clang: numpy/core/src/multiarray/buffer.c
clang: src/multiarray/cblasfuncs.c
clang: error: no such file or directory: 'src/multiarray/cblasfuncs.c'
clang: error: no input files
clang: error: no such file or directory: 'src/multiarray/cblasfuncs.c'
clang: error: no input files
Indeed, that file is not present in the tarball. My download of the
tarball from sourceforge has sha1
hash 424bcee49507a655260068e767bd198094fc5604.
It looks like the tarball from github (
https://github.com/numpy/numpy/archive/v1.10.0b1.tar.gz) has the needed
file.
Hmm, interesting. I'll take a look.
I'm uploading replacements for the tar and zip files.. Not sure why
numpy-vendor didn't do the job.

Chuck
Sandro Tosi
2015-08-11 22:32:47 UTC
Permalink
On Tue, Aug 11, 2015 at 11:22 PM, Charles R Harris
Post by Charles R Harris
I'm uploading replacements for the tar and zip files.. Not sure why
numpy-vendor didn't do the job.
if so then please upload a b2, so it will avoid confusion with those
who have downloaded a b1-pre-fix and those with a b1-post-fix

Regards,
--
Sandro Tosi (aka morph, morpheus, matrixhasu)
My website: http://matrixhasu.altervista.org/
Me at Debian: http://wiki.debian.org/SandroTosi
Charles R Harris
2015-08-11 22:46:31 UTC
Permalink
Post by Sandro Tosi
On Tue, Aug 11, 2015 at 11:22 PM, Charles R Harris
Post by Charles R Harris
I'm uploading replacements for the tar and zip files.. Not sure why
numpy-vendor didn't do the job.
if so then please upload a b2, so it will avoid confusion with those
who have downloaded a b1-pre-fix and those with a b1-post-fix
I think it was caught early enough that it won't be much of a problem. I
expect there will be other problems that will require a beta 2 soon enough
;)

Chuck
Sandro Tosi
2015-08-12 00:04:53 UTC
Permalink
On Tue, Aug 11, 2015 at 11:46 PM, Charles R Harris
Post by Charles R Harris
Post by Sandro Tosi
On Tue, Aug 11, 2015 at 11:22 PM, Charles R Harris
Post by Charles R Harris
I'm uploading replacements for the tar and zip files.. Not sure why
numpy-vendor didn't do the job.
if so then please upload a b2, so it will avoid confusion with those
who have downloaded a b1-pre-fix and those with a b1-post-fix
I think it was caught early enough that it won't be much of a problem. I
expect there will be other problems that will require a beta 2 soon enough
;)
we can agree to disagree, but nevermind :) I gave it a quick build
test on debian unstable amd64 and it's building correctly, I will
finalize the package in the next few days and upload to the various
weird architectures Debian supports and see where it fails.

Regards,
--
Sandro Tosi (aka morph, morpheus, matrixhasu)
My website: http://matrixhasu.altervista.org/
Me at Debian: http://wiki.debian.org/SandroTosi
Jens Jørgen Mortensen
2015-08-12 07:41:57 UTC
Permalink
Post by Charles R Harris
Hi All,
give this release a whirl and report any problems either on the
numpy-discussion list or by opening an issue on github.
I'm pleased to announce the first beta release of Numpy 1.10.0. There
is over a year's worth of enhancements and bug fixes in the 1.10.0
release, so please give this release a whirl and report any problems
either on the numpy-discussion list or by opening an issue on github.
Tarballs, installers, and release notes may be found in the usual
place at Sourceforge
<https://sourceforge.net/projects/numpy/files/NumPy/1.10.0b1/>.
This looks a bit strange:

Python 2.7.9 (default, Apr 2 2015, 15:33:21)
[GCC 4.9.2] on linux2
Type "help", "copyright", "credits" or "license" for more information.
Post by Charles R Harris
import numpy as np
np.zeros(1).strides
(9223372036854775807,)
Post by Charles R Harris
np.zeros(42).strides
(8,)
Post by Charles R Harris
np.__version__
'1.10.0b1'

This is on Ubuntu 15.04.

Jens Jørgen
Post by Charles R Harris
Chuck
_______________________________________________
SciPy-Dev mailing list
http://mail.scipy.org/mailman/listinfo/scipy-dev
Sebastian Berg
2015-08-12 07:51:39 UTC
Permalink
Post by Jens Jørgen Mortensen
Post by Charles R Harris
Hi All,
give this release a whirl and report any problems either on the
numpy-discussion list or by opening an issue on github.
I'm pleased to announce the first beta release of Numpy 1.10.0.
There is over a year's worth of enhancements and bug fixes in the
1.10.0 release, so please give this release a whirl and report any
problems either on the numpy-discussion list or by opening an issue
on github. Tarballs, installers, and release notes may be found in
the usual place at Sourceforge.
It is intentional, it will not be the case in the final release.

And thanks Chuck for all the release work!

- Sebastian
Post by Jens Jørgen Mortensen
Python 2.7.9 (default, Apr 2 2015, 15:33:21)
[GCC 4.9.2] on linux2
Type "help", "copyright", "credits" or "license" for more information.
Post by Charles R Harris
import numpy as np
np.zeros(1).strides
(9223372036854775807,)
Post by Charles R Harris
np.zeros(42).strides
(8,)
Post by Charles R Harris
np.__version__
'1.10.0b1'
This is on Ubuntu 15.04.
Jens JÞrgen
Post by Charles R Harris
Chuck
_______________________________________________
SciPy-Dev mailing list
http://mail.scipy.org/mailman/listinfo/scipy-dev
_______________________________________________
NumPy-Discussion mailing list
http://mail.scipy.org/mailman/listinfo/numpy-discussion
Nathaniel Smith
2015-08-12 08:07:37 UTC
Permalink
On Wed, Aug 12, 2015 at 12:51 AM, Sebastian Berg
Post by Sebastian Berg
Post by Jens Jørgen Mortensen
Post by Charles R Harris
Hi All,
give this release a whirl and report any problems either on the
numpy-discussion list or by opening an issue on github.
I'm pleased to announce the first beta release of Numpy 1.10.0.
There is over a year's worth of enhancements and bug fixes in the
1.10.0 release, so please give this release a whirl and report any
problems either on the numpy-discussion list or by opening an issue
on github. Tarballs, installers, and release notes may be found in
the usual place at Sourceforge.
Python 2.7.9 (default, Apr 2 2015, 15:33:21)
[GCC 4.9.2] on linux2
Type "help", "copyright", "credits" or "license" for more information.
Post by Charles R Harris
import numpy as np
np.zeros(1).strides
(9223372036854775807,)
Post by Charles R Harris
np.zeros(42).strides
(8,)
Post by Charles R Harris
np.__version__
'1.10.0b1'
It is intentional, it will not be the case in the final release.
Given how quickly this surprised someone, it looks like it would be
helpful to have some single link we could give people to explain
what's going on here. Do we have such a thing? In a few minutes of
searching all I was able to find was

http://docs.scipy.org/doc/numpy/release.html#npy-relaxed-strides-checking
https://github.com/numpy/numpy/blob/master/doc/release/1.10.0-notes.rst#relaxed-stride-checking

which together kinda sorta hint at what's going on if you squint, but
not really? Maybe we should add a paragraph to the 1.10 release notes?
Post by Sebastian Berg
And thanks Chuck for all the release work!
Indeed!

-n
--
Nathaniel J. Smith -- http://vorpus.org
Sebastian Berg
2015-08-12 11:23:46 UTC
Permalink
Post by Nathaniel Smith
On Wed, Aug 12, 2015 at 12:51 AM, Sebastian Berg
Post by Sebastian Berg
Post by Jens Jørgen Mortensen
Post by Charles R Harris
Hi All,
give this release a whirl and report any problems either on the
numpy-discussion list or by opening an issue on github.
I'm pleased to announce the first beta release of Numpy 1.10.0.
There is over a year's worth of enhancements and bug fixes in the
1.10.0 release, so please give this release a whirl and report any
problems either on the numpy-discussion list or by opening an issue
on github. Tarballs, installers, and release notes may be found in
the usual place at Sourceforge.
Python 2.7.9 (default, Apr 2 2015, 15:33:21)
[GCC 4.9.2] on linux2
Type "help", "copyright", "credits" or "license" for more information.
Post by Charles R Harris
import numpy as np
np.zeros(1).strides
(9223372036854775807,)
Post by Charles R Harris
np.zeros(42).strides
(8,)
Post by Charles R Harris
np.__version__
'1.10.0b1'
It is intentional, it will not be the case in the final release.
Given how quickly this surprised someone, it looks like it would be
helpful to have some single link we could give people to explain
what's going on here. Do we have such a thing? In a few minutes of
searching all I was able to find was
http://docs.scipy.org/doc/numpy/release.html#npy-relaxed-strides-checking
https://github.com/numpy/numpy/blob/master/doc/release/1.10.0-notes.rst#relaxed-stride-checking
which together kinda sorta hint at what's going on if you squint, but
not really? Maybe we should add a paragraph to the 1.10 release notes?
True, frankly, after I hit send I thought I should have explained more
in any case.
I think relaxed strides is explained (though 1.10 could possible
link/include more). The issue is that I/we forgot to mention the "funny"
stride messing up to expose bugs/help debugging....

So in case someone wonders. When relaxed strides is active, we
intentionally give this funny stride that Jens saw (we will not do this
in a final release) because failure to work correctly with it hints to
general bugs and tests are likely to miss without this "help". Of course
some software that is totally fine might still stumble on these strides,
though I expect making it work fine with them should not be hard and
make it more robust in any case.

- Sebastian
Post by Nathaniel Smith
Post by Sebastian Berg
And thanks Chuck for all the release work!
Indeed!
-n
Matthew Brett
2015-08-13 12:04:34 UTC
Permalink
Hi,

On Wed, Aug 12, 2015 at 12:23 PM, Sebastian Berg
Post by Nathaniel Smith
On Wed, Aug 12, 2015 at 12:51 AM, Sebastian Berg
Post by Charles R Harris
Hi All,
give this release a whirl and report any problems either on the
numpy-discussion list or by opening an issue on github.
I'm pleased to announce the first beta release of Numpy 1.10.0.
There is over a year's worth of enhancements and bug fixes in the
1.10.0 release, so please give this release a whirl and report any
problems either on the numpy-discussion list or by opening an issue
on github. Tarballs, installers, and release notes may be found in
the usual place at Sourceforge.
I'm getting test errors on the standard OSX numpy / scipy compilation rig:

Python.org Python
OSX 10.9
clang
gfortran 4.2.3
Compiling from the `maintenance/1.10.x` branch (is there a 1.10.0b1 tag)?

======================================================================
ERROR: test_accelerate_framework_sgemv_fix (test_multiarray.TestDot)
----------------------------------------------------------------------
Traceback (most recent call last):
File "/Users/mb312/.virtualenvs/test/lib/python2.7/site-packages/numpy/core/tests/test_multiarray.py",
line 4218, in test_accelerate_framework_sgemv_fix
m = aligned_array(100, 15, np.float32)
File "/Users/mb312/.virtualenvs/test/lib/python2.7/site-packages/numpy/core/tests/test_multiarray.py",
line 4200, in aligned_array
d = np.dtype()
TypeError: Required argument 'dtype' (pos 1) not found

This one should be fixed by https://github.com/numpy/numpy/pull/6202

======================================================================
ERROR: test_callback.TestF77Callback.test_string_callback
----------------------------------------------------------------------
Traceback (most recent call last):
File "/Users/mb312/.virtualenvs/test/lib/python2.7/site-packages/nose/case.py",
line 381, in setUp
try_run(self.inst, ('setup', 'setUp'))
File "/Users/mb312/.virtualenvs/test/lib/python2.7/site-packages/nose/util.py",
line 471, in try_run
return func()
File "/Users/mb312/.virtualenvs/test/lib/python2.7/site-packages/numpy/f2py/tests/util.py",
line 362, in setUp
module_name=self.module_name)
File "/Users/mb312/.virtualenvs/test/lib/python2.7/site-packages/numpy/f2py/tests/util.py",
line 79, in wrapper
memo[key] = func(*a, **kw)
File "/Users/mb312/.virtualenvs/test/lib/python2.7/site-packages/numpy/f2py/tests/util.py",
line 170, in build_code
module_name=module_name)
File "/Users/mb312/.virtualenvs/test/lib/python2.7/site-packages/numpy/f2py/tests/util.py",
line 79, in wrapper
memo[key] = func(*a, **kw)
File "/Users/mb312/.virtualenvs/test/lib/python2.7/site-packages/numpy/f2py/tests/util.py",
line 150, in build_module
__import__(module_name)
ImportError: dlopen(/var/folders/s7/r25pn2xj48n4cm76_mgsb78h0000gn/T/tmpa39XPB/_test_ext_module_5403.so,
2): Symbol not found: _func0_
Referenced from:
/var/folders/s7/r25pn2xj48n4cm76_mgsb78h0000gn/T/tmpa39XPB/_test_ext_module_5403.so
Expected in: dynamic lookup

Any ideas about this second one?

Cheers,

Matthew
Matthew Brett
2015-08-13 22:34:15 UTC
Permalink
Post by Matthew Brett
Hi,
On Wed, Aug 12, 2015 at 12:23 PM, Sebastian Berg
Post by Nathaniel Smith
On Wed, Aug 12, 2015 at 12:51 AM, Sebastian Berg
Post by Charles R Harris
Hi All,
give this release a whirl and report any problems either on the
numpy-discussion list or by opening an issue on github.
I'm pleased to announce the first beta release of Numpy 1.10.0.
There is over a year's worth of enhancements and bug fixes in the
1.10.0 release, so please give this release a whirl and report any
problems either on the numpy-discussion list or by opening an issue
on github. Tarballs, installers, and release notes may be found in
the usual place at Sourceforge.
Python.org Python
OSX 10.9
clang
gfortran 4.2.3
Compiling from the `maintenance/1.10.x` branch (is there a 1.10.0b1 tag)?
======================================================================
ERROR: test_accelerate_framework_sgemv_fix (test_multiarray.TestDot)
----------------------------------------------------------------------
File "/Users/mb312/.virtualenvs/test/lib/python2.7/site-packages/numpy/core/tests/test_multiarray.py",
line 4218, in test_accelerate_framework_sgemv_fix
m = aligned_array(100, 15, np.float32)
File "/Users/mb312/.virtualenvs/test/lib/python2.7/site-packages/numpy/core/tests/test_multiarray.py",
line 4200, in aligned_array
d = np.dtype()
TypeError: Required argument 'dtype' (pos 1) not found
This one should be fixed by https://github.com/numpy/numpy/pull/6202
======================================================================
ERROR: test_callback.TestF77Callback.test_string_callback
----------------------------------------------------------------------
File "/Users/mb312/.virtualenvs/test/lib/python2.7/site-packages/nose/case.py",
line 381, in setUp
try_run(self.inst, ('setup', 'setUp'))
File "/Users/mb312/.virtualenvs/test/lib/python2.7/site-packages/nose/util.py",
line 471, in try_run
return func()
File "/Users/mb312/.virtualenvs/test/lib/python2.7/site-packages/numpy/f2py/tests/util.py",
line 362, in setUp
module_name=self.module_name)
File "/Users/mb312/.virtualenvs/test/lib/python2.7/site-packages/numpy/f2py/tests/util.py",
line 79, in wrapper
memo[key] = func(*a, **kw)
File "/Users/mb312/.virtualenvs/test/lib/python2.7/site-packages/numpy/f2py/tests/util.py",
line 170, in build_code
module_name=module_name)
File "/Users/mb312/.virtualenvs/test/lib/python2.7/site-packages/numpy/f2py/tests/util.py",
line 79, in wrapper
memo[key] = func(*a, **kw)
File "/Users/mb312/.virtualenvs/test/lib/python2.7/site-packages/numpy/f2py/tests/util.py",
line 150, in build_module
__import__(module_name)
ImportError: dlopen(/var/folders/s7/r25pn2xj48n4cm76_mgsb78h0000gn/T/tmpa39XPB/_test_ext_module_5403.so,
2): Symbol not found: _func0_
/var/folders/s7/r25pn2xj48n4cm76_mgsb78h0000gn/T/tmpa39XPB/_test_ext_module_5403.so
Expected in: dynamic lookup
Any ideas about this second one?
I don't get this second error when building with homebrew gfortran
4.8. Is this expected? Should we be raising an error for earlier
gfortrans?

Cheers,

Matthew
Charles R Harris
2015-08-13 23:37:27 UTC
Permalink
Post by Matthew Brett
Post by Matthew Brett
Hi,
On Wed, Aug 12, 2015 at 12:23 PM, Sebastian Berg
Post by Nathaniel Smith
On Wed, Aug 12, 2015 at 12:51 AM, Sebastian Berg
Post by Charles R Harris
Hi All,
give this release a whirl and report any problems either on the
numpy-discussion list or by opening an issue on github.
I'm pleased to announce the first beta release of Numpy 1.10.0.
There is over a year's worth of enhancements and bug fixes in the
1.10.0 release, so please give this release a whirl and report any
problems either on the numpy-discussion list or by opening an
issue
Post by Matthew Brett
Post by Nathaniel Smith
Post by Charles R Harris
on github. Tarballs, installers, and release notes may be found in
the usual place at Sourceforge.
I'm getting test errors on the standard OSX numpy / scipy compilation
Python.org Python
OSX 10.9
clang
gfortran 4.2.3
Compiling from the `maintenance/1.10.x` branch (is there a 1.10.0b1 tag)?
======================================================================
ERROR: test_accelerate_framework_sgemv_fix (test_multiarray.TestDot)
----------------------------------------------------------------------
File
"/Users/mb312/.virtualenvs/test/lib/python2.7/site-packages/numpy/core/tests/test_multiarray.py",
Post by Matthew Brett
line 4218, in test_accelerate_framework_sgemv_fix
m = aligned_array(100, 15, np.float32)
File
"/Users/mb312/.virtualenvs/test/lib/python2.7/site-packages/numpy/core/tests/test_multiarray.py",
Post by Matthew Brett
line 4200, in aligned_array
d = np.dtype()
TypeError: Required argument 'dtype' (pos 1) not found
This one should be fixed by https://github.com/numpy/numpy/pull/6202
======================================================================
ERROR: test_callback.TestF77Callback.test_string_callback
----------------------------------------------------------------------
File
"/Users/mb312/.virtualenvs/test/lib/python2.7/site-packages/nose/case.py",
Post by Matthew Brett
line 381, in setUp
try_run(self.inst, ('setup', 'setUp'))
File
"/Users/mb312/.virtualenvs/test/lib/python2.7/site-packages/nose/util.py",
Post by Matthew Brett
line 471, in try_run
return func()
File
"/Users/mb312/.virtualenvs/test/lib/python2.7/site-packages/numpy/f2py/tests/util.py",
Post by Matthew Brett
line 362, in setUp
module_name=self.module_name)
File
"/Users/mb312/.virtualenvs/test/lib/python2.7/site-packages/numpy/f2py/tests/util.py",
Post by Matthew Brett
line 79, in wrapper
memo[key] = func(*a, **kw)
File
"/Users/mb312/.virtualenvs/test/lib/python2.7/site-packages/numpy/f2py/tests/util.py",
Post by Matthew Brett
line 170, in build_code
module_name=module_name)
File
"/Users/mb312/.virtualenvs/test/lib/python2.7/site-packages/numpy/f2py/tests/util.py",
Post by Matthew Brett
line 79, in wrapper
memo[key] = func(*a, **kw)
File
"/Users/mb312/.virtualenvs/test/lib/python2.7/site-packages/numpy/f2py/tests/util.py",
Post by Matthew Brett
line 150, in build_module
__import__(module_name)
dlopen(/var/folders/s7/r25pn2xj48n4cm76_mgsb78h0000gn/T/tmpa39XPB/_test_ext_module_5403.so,
Post by Matthew Brett
2): Symbol not found: _func0_
/var/folders/s7/r25pn2xj48n4cm76_mgsb78h0000gn/T/tmpa39XPB/_test_ext_module_5403.so
Post by Matthew Brett
Expected in: dynamic lookup
Any ideas about this second one?
I don't get this second error when building with homebrew gfortran
4.8. Is this expected? Should we be raising an error for earlier
gfortrans?
Probaby, or we could make the failing bit, if we can find it, gcc version
dependent. gcc 4.2 is eight years old, which OS X versions depend on it?

Chuck

Loading...